Hm, those settings weren't as well-behaved as I thought. I wonder if there is some unpleasant randomness here

This commit is contained in:
Chris Cannam
2020-10-22 17:13:55 +01:00
parent e3e5fe7ae3
commit dbaaf12598

View File

@@ -155,10 +155,9 @@ D_IPP::D_IPP(Resampler::Quality /* quality */,
cerr << "Resampler::Resampler: using IPP implementation" << endl; cerr << "Resampler::Resampler: using IPP implementation" << endl;
} }
int nStep = 16; m_window = 32;
int nStep = 64;
IppHintAlgorithm hint = ippAlgHintFast; IppHintAlgorithm hint = ippAlgHintFast;
m_window = 48;
nStep = 80;
m_factor = 8; // initial upper bound on m_ratio, may be amended later m_factor = 8; // initial upper bound on m_ratio, may be amended later
// This is largely based on the IPP docs and examples. Adapted // This is largely based on the IPP docs and examples. Adapted